Informatics, Volume 10, Issue 2 (June 2023) – 22 articles

Risk management is a highly important issue for Fintech companies; moreover, it is very specific and puts forward the serious requirements toward the top management of any financial institution. This study was devoted to specifying the risk factors affecting the finance and capital adequacy of financial institutions. The authors considered the different types of risks in combination, whereas other scholars usually analyze risks in isolation; however, the authors believe that it is necessary to consider their mutual impact. The risks were estimated using the PLS-SEM method in Smart PLS-4 software. The quality of the obtained model is very high according to all indicators. Five hypotheses related to finance and five hypotheses related to capital adequacy were considered. The impact of AML, cyber, and governance risks on capital adequacy was confirmed; the effect of governance and operational risks on finance was also confirmed. Other risks have no impact on finance and capital adequacy. It is interesting that risks associated with staff have no impact on finance and capital adequacy. The findings of this study can be easily applied by any financial institution for risk analysis. Moreover, this study can serve toward a better collaboration of scholars investigating the Fintech activities and practitioners working in this sphere. The authors present a novel approach for enhancing key performance indicators (KPIs) for Fintech companies, proposing utilizing metrics that are derived from the company’s specific risks, thereby introducing an innovative method for selecting KPIs based on the inherent risks associated with the Fintech’s business model. This model aligns the KPIs with the unique risk profile of the company, fostering a fresh perspective on performance measurement within the Fintech industry. Full article

One way to improve Indonesia’s ranking in terms of ease of conducting business is by taking a closer look at the business licensing process. This study aims to carry out an assessment using a smart governance framework and recommendation capabilities from the Enterprise System (ES). As a result, the recommendations for improvement with the expected priority are generated. The stages of this research are observing the process of making bio-business permits, followed by interviews related to several Enterprise Architecture (EA) capabilities, and providing recommendations based on the results of the maturity level of IT governance. These recommendations are then mapped into an impact—effort matrix for program prioritization. The recommendations for bio-business licenses can also be used to improve the process for other business licenses. Implementation of the EA framework has been proven to align technology, organization, and processes so that it can support continuous improvement processes. Full article

Using camera-based algorithms to detect abnormal patterns in children’s handwriting has become a promising tool in education and occupational therapy. This study analyzes the performance of a camera- and tablet-based handwriting verification algorithm to detect abnormal patterns in handwriting samples processed from 71 students of different grades. The study results revealed that the algorithm saw abnormal patterns in 20% of the handwriting samples processed, which included practices such as delayed typing speed, excessive pen pressure, irregular slant, and lack of word spacing. In addition, it was observed that the detection accuracy of the algorithm was 95% when comparing the camera data with the abnormal patterns detected, which indicates a high reliability in the results obtained. The highlight of the study was the feedback provided to children and teachers on the camera data and any abnormal patterns detected. This can significantly impact students’ awareness and improvement of writing skills by providing real-time feedback on their writing and allowing them to adjust to correct detected abnormal patterns. Full article

In computer vision, ethnicity classification tasks utilize images containing human faces to extract ethnicity labels. Ethnicity is one of the soft biometric feature categories useful in data analysis for commercial, public, and health sectors. Ethnicity classification begins with face detection as a preprocessing process to determine a human’s presence; then, the feature representation is extracted from the isolated facial image to predict the ethnicity class. This study utilized four handcrafted features (multi-local binary pattern (MLBP), histogram of gradient (HOG), color histogram, and speeded-up-robust-features-based (SURF-based)) as the basis for the generation of a compact-fusion feature. The compact-fusion framework involves optimal feature selection, compact feature extraction, and compact-fusion feature representation. The final feature representation was trained and tested with the SVM One Versus All classifier for ethnicity classification. When it was evaluated in two large datasets, UTKFace and Fair Face, the proposed framework achieved accuracy levels of 89.14%, 82.19%, and 73.87%, respectively, for the UTKFace dataset with four or five classes and the Fair Face dataset with four classes. Furthermore, the compact-fusion feature with a small number of features at 4790, constructed based on conventional handcrafted features, achieved competitive results compared with state-of-the-art methods using a deep-learning-based approach. Full article

In the context of developing machine learning models, until and unless we have the required data engineering and machine learning development competencies as well as the time to train and test different machine learning models and tune their hyperparameters, it is worth trying out the automatic machine learning features provided by several cloud-based and cloud-agnostic platforms. This paper explores the possibility of generating automatic machine learning models with low-code experience. We developed criteria to compare different machine learning platforms for generating automatic machine learning models and presenting their results. Thereafter, lessons learned by developing automatic machine learning models from a sample dataset across four different machine learning platforms were elucidated. We also interviewed machine learning experts to conceptualize their domain-specific problems that automatic machine learning platforms can address. Results showed that automatic machine learning platforms can provide a fast track for organizations seeking the digitalization of their businesses. Automatic machine learning platforms help produce results, especially for time-constrained projects where resources are lacking. The contribution of this paper is in the form of a lab experiment in which we demonstrate how low-code platforms can provide a viable option to many business cases and, henceforth, provide a lane that is faster than the usual hiring and training of already scarce data scientists and to analytics projects that suffer from overruns. Full article

Unsuitable shoe shapes and sizes are a critical reason for unhealthy feet, may severely contribute to chronic injuries such as foot ulcers in susceptible people (e.g., diabetes patients), and thus need accurate measurements in the manner of expert-based procedures. However, the manual measure of such accurate shapes and sizes is labor-intensive, time-consuming, and impractical to apply in a real-time system. This research proposes a foot-detection approach using expert-like measurements to address this concern. It combines the seven-foot dimensions model and the light detection and ranging sensor to encode foot shapes and sizes and detect the dimension surfaces. The graph-based algorithms are developed to present seven-foot dimensions and visualize the shoe’s model based on the augmented reality (AR) technique. The results show that our approach can detect shapes and sizes more effectively than the traditional approach, helps the system imitate expert-like measurements accurately, and can be employed in intelligent applications for susceptible people-based feet measurements. Full article

We were promised that the Metaverse would revolutionize our lives, social interactions, work, and business. However, how and when will this happen? We have seen the growth and development of technology, but there is no agreement or prediction about a specific time, and we can only follow the how question. To investigate more leads about this concept, we considered a main research question: How is the Metaverse actually being perceived? This question is connected with three objectives: to verify how the Metaverse is being represented and characterized, identify the main dimensions that facilitate or influence the acceptance of the Metaverse, and identify the leading technologies that suit the Metaverse concept. This study consisted of a documental analysis—or meta-analysis—of fifty of the most relevant scientific papers (taking into account some inclusion criteria) published in the last three years, using the Leximancer software to create concept maps to illustrate the main concepts and themes extracted from the articles to understand their associations or relations with the Metaverse concept. This study provided us with essential findings about how this concept has been perceived and allowed us to answer our objectives, contributing to a scientific discussion on the topic, and provided some valid suggestions for future research, which is already in progress. It also provided new leads on approaching this concept in development. Full article

Alzheimer’s disease (AD) and AD-related dementias (AD/ADRD) are a group of progressive neurodegenerative diseases. The progression of AD can be conceptualized as a continuum in which patients progress from normal cognition to preclinical AD (i.e., no symptoms but biological changes in the brain) to mild cognitive impairment (MCI) due to AD (i.e., mild symptoms but not interfere with daily activities), followed by increasing severity of dementia due to AD. Early detection and prediction models for the transition of MCI to AD/ADRD are needed, and efforts have been made to build predictions of MCI conversion to AD/ADRD. However, most existing studies developing such prediction models did not consider the competing risks of death, which may result in biased risk estimates. In this study, we aim to develop a prediction model for AD/ADRD among patients with MCI considering the competing risks of death using a semi-competing risk approach. Full article

During the last few years, learning techniques have changed, both in basic education and in higher education. This change has been accompanied by new technologies such as Augmented Reality (AR) and Virtual Reality (AR). The combination of these technologies in education has allowed a greater immersion, positively affecting the learning and teaching processes. In addition, since the COVID-19 pandemic, this trend has been growing due to the diversity of the different fields of application of these technologies, such as heterogeneity in their combination and their different experiences. It is necessary to review the state of the art to determine the effectiveness of the application of these technologies in the field of university higher education. In the present paper, this aim is achieved by performing a systematic literature review from 2012 to 2022. A total of 129 papers were analyzed. Studies in our review concluded that the application of AR/VR improves learning immersion, especially in hospitality, medicine, and science studies. However, there are also negative effects of using these technologies, such as visual exhaustion and mental fatigue. Full article

In Indonesia, there is still a disparity in telecommunications access, with most rural areas experiencing “no signal” or “blank spots.” In contrast, urban areas enjoy modern and societally-beneficial technologies. A comprehensive framework is needed to address the disparity in telecommunications access between “rich” and “poor” groups in urban and rural/remote areas, respectively. This paper proposes a framework, built by the mathematical model, that can be used as a reference for the Indonesian government in constructing the nation’s telecommunications infrastructure. The framework categorizes Indonesian administrative regions into four grids: Grid #1: “fostered” districts; Grid #2: “developing” districts; Grid #3: “developed” districts; and Grid #4: “independent-advanced” districts. To determine where each district falls in these grids, we propose a novel statistical approach using 17 indicators involving a telecommunications network and socioeconomic factors. The proposed framework results in a grid visualization of 7232 districts in Indonesia. Finally, as this paper is replete with academic research approaches and mathematical model perspectives, it is expected that the results may be a valuable input to the development of the country’s telecommunications policy. Full article

Studies have explored the factors influencing the safety of PTWs; however, very little has been carried out to comprehensively investigate the factors influencing fatal PTW crashes while considering the fault status of the rider in crash hotspot areas. This study employs spatio-temporal hotspot analysis and association rule mining techniques to discover hidden associations between crash risk factors that lead to fatal PTW crashes considering the fault status of the rider at statistically significant PTW crash hotspots in South Korea from 2012 to 2017. The results indicate the presence of consecutively fatal PTW crash hotspots concentrated within Korea’s densely populated capital, Seoul, and new hotspots near its periphery. According to the results, violations such as over-speeding and red-light running were critical contributory factors influencing PTW crashes at hotspots during summer and at intersections. Interestingly, while reckless riding was the main traffic violation leading to PTW rider at-fault crashes at hotspots, violations such as improper safety distance and red-light running were strongly associated with PTW rider not-at-fault crashes at hotspots. In addition, while PTW rider at-fault crashes are likely to occur during summer, PTW rider not-at-fault crashes mostly occur during spring. The findings could be used for developing targeted policies for improving PTW safety at hotspots. Full article

The Jewish community archive in Pisa owns a vast collection of documents and manuscripts that date back centuries. These documents contain valuable genealogical information, including birth, marriage, and death records. This paper aims to describe the preliminary results of the Archivio Storico della Comunita Ebraica di Pisa (ASCEPI) project, with a focus on the extraction of data from the Nati, Morti e Ballottati (NMB) Registry document in the archive. The NMB Registry contains about 1900 records of births, deaths, and balloted individuals within the Jewish community in Pisa. The study uses a semiautomatic pipeline of digitization, transcription, and Natural Language Processing (NLP) techniques to extract personal data such as names, surnames, birth and death dates, and parental names from each record. The extracted data are then used to build a knowledge base and a genealogical tree for a representative family, Supino. This study demonstrates the potential of using NLP and rule-based techniques to extract valuable information from historical documents and to construct genealogical trees. Full article

Digital platform business model innovation is a rapidly evolving field, yet the literature on resource, complementary, and ecological boundaries remains limited, leaving a significant gap in our understanding of the factors that shape the success of these platforms. This paper explores the mechanisms by which digital platforms enable business model innovation, a topic of significant theoretical and practical importance that has yet to be fully examined. Through a review of the existing literature and an examination of the connotations of digital platforms, the design of platform boundaries, and the deployment of boundary resources, the study finds that (1) the uncertainty of complementors and complementary products drives business model innovation in digital platforms; (2) the design of resource, complementary, and ecological system boundaries is crucial to digital platform business models and manages complementor and complementary product uncertainty while promoting value co-creation; and (3) boundary resources establish, manage, and sustain cross-border relationships that impact value creation and capture. Based on these findings, four research propositions are proposed to guide future research on digital platform business model innovation and provide insights for effectively innovating business models and influencing value creation and capture. Full article

This paper provides new models of the attention-based random forests called LARF (leaf attention-based random forest). The first idea behind the models is to introduce a two-level attention, where one of the levels is the “leaf” attention, and the attention mechanism is applied to every leaf of trees. The second level is the tree attention depending on the “leaf” attention. The second idea is to replace the softmax operation in the attention with the weighted sum of the softmax operations with different parameters. It is implemented by applying a mixture of Huber’s contamination models and can be regarded as an analog of the multi-head attention, with “heads” defined by selecting a value of the softmax parameter. Attention parameters are simply trained by solving the quadratic optimization problem. To simplify the tuning process of the models, it is proposed to convert the tuning contamination parameters into trainable parameters and to compute them by solving the quadratic optimization problem. Many numerical experiments with real datasets are performed for studying LARFs. The code of the proposed algorithms is available. Full article

There are positives and negatives of using YouTube in terms of loneliness and mental health. YouTube’s streaming content is an amazing resource, however, there may be bias or errors in its recommendation algorithms. Parasocial relationships can also complicate the impact of YouTube use. Intervention may be necessary when problematic and risky content is associated with unhealthy behaviors and negative impacts on mental health. Children and adolescents are particularly vulnerable. Although YouTube might assist in connecting with peers, there are privacy, safety, and quality issues to consider. This paper is an integrative review of the positive and negative impacts of YouTube with the aim to inform the design and development of a technology-based intervention to improve mental health. The impact of YouTube use on loneliness and mental health was explored by synthesizing a purposive selection (n = 32) of the empirical and theoretical literature. Next, we explored human–computer interaction issues and proposed a concept whereby an independent-of-YouTube algorithmic recommendation system steers users toward verified positive mental health content or promotions. Full article

The COVID-19 pandemic and the boom of fake news cluttering the internet have revealed the power of social media today. However, young people are not yet aware of their role in the digital age, even though they are the main users of social media. As a result, the belief that older adults are responsible for information is being re-evaluated. In light of this, the present study was aimed at identifying the factors associated with the spread of fake news among young people in Medellín (Colombia). A total of 404 self-administered questionnaires were processed in a sample of people between the ages of 18 and 34 and analyzed using statistical techniques, such as exploratory factor analysis and structural equation modeling. The results suggest that the instantaneous sharing of fake news is linked to people’s desire to raise awareness among their inner circle, particularly when the messages shared are consistent with their perceptions and beliefs, or to the lack of time to properly verify their accuracy. Finally, passive corrective actions were found to have a less significant impact in the Colombian context than in the context of the original model, which may be explained by cultural factors. Full article

During the pandemic, universities were forced to convert their educational process online. Students had to adapt to new educational conditions and the proposed online environment. Now, we are back to the traditional blended learning environment and wish to understand the students’ attitudes and perceptions of online learning, knowing that they are able to compare blended and online modes. The aim of this paper is to present the performed predictive analysis regarding the students’ online learning performance taking into account their opinion. The predictive models are created through a supervised machine learning algorithm based on Artificial Neural Networks and are characterized with high accuracy. The analysis is based on generated synthetic datasets, ensuring a high level of students’ privacy preservation. Full article

Online commerce has been growing rapidly in an increasingly digital world, and gamification, the practice of designing games in a context outside the industry itself, can be an effective strategy to stimulate consumer engagement and conversion rate. This paper describes the design process involved in introducing gamification into an online shop that is supported by two game servers of the same kind, namely one in the United States of America (US) and another in Portugal (PT). Through the various phases of the design thinking process, a gamified system was implemented to meet the needs of various types of users frequently found in the shops. The gamification elements used were intended to increase user engagement with the shops so that they would become more aware of existing products and the introduction of new products, promoting purchase through intangible challenges and rewards. The impacts on server revenues and user satisfaction (N = 138) were evaluated one month after introducing the gamification techniques. The results show that gamification has a positive effect on users, with a significant increase in consumer interaction in both shops. However, from a business point of view, the results show only an increase in revenue for the US shop, while the Portuguese shop shows no significant differences compared to previous months. Of the two user groups analyzed, only those who frequent the US shop show receptivity toward intangible rewards, with tangible rewards (discounts) being a greater motivating factor for both groups. Full article

Jordanian healthcare institutes have launched several programs since 2009 to establish health information systems (HISs). Nowadays, the generic expectation is that the use of HIS resources is performed on daily basis among healthcare staff. However, there can be still a noticeable barrier due to a lack of knowledge if medical doctors do not receive proper training on existing HISs. Moreover, the lack of studies on this area hinders the clarity about the received versus the required training skills among medical doctors. To support this research initiative, survey data have been collected from specialized medical doctors who are currently affiliated with five Jordanian universities to assess their need for HIS training. The results also aim to explore the extent of medical doctors’ use of HIS resources in Jordan. Moreover, they examine whether medical doctors require additional training on using HIS resources or not, as well as the main areas of required training programs. Specifically, this paper highlights the main topics that can be suitable subjects for enhanced training programs. The results show that most respondents use HISs in their daily clinical practices. However, most of them have not taken professional training on such systems. Hence, most of the respondents reported the need for additional training programs on several aspects of HIS resources. Moreover, based on the survey results, the most significant areas that require training are biomedical data analysis, artificial intelligence in medicine, health care management, and recent advances in electronic health records, respectively. Therefore, specialized medical doctors in Jordan need training on extracting useful and potential features of HISs. Education and training professionals in healthcare are recommended to establish training programs in Jordanian healthcare centers, which can further improve the quality of healthcare. Full article

The paraphrase generator for citation sentences is used to produce several sentence alternatives to avoid plagiarism. Furthermore, the generation results need to pay attention to semantic similarity and lexical divergence standards. This study proposed the StoPGEN model as an algorithm for generating citation paraphrase sentences with stochastic output. The generation process is guided by an objective function using a simulated annealing algorithm to maintain the properties of semantic similarity and lexical divergence. The objective function is created by combining the two factors that maintain these properties. This study combined METEOR and PINC Scores in a linear weighting function that can be adjusted for its value tendency in one of the matrix functions. The dataset of citation sentences that had been labeled with paraphrases was used to test StoPGEN and other models for comparison. The StoPGEN model, with the citation sentences dataset, produced a BLEU score of 55.37, outperforming the bidirectional LSTM method with a value of 28.93. StoPGEN was also tested using Quora data by changing the language source in the architecture section resulting in a BLEU score of 22.37, outperforming UPSA 18.21. In addition, the qualitative evaluation results of the citation sentence generation based on respondents obtained an acceptance value of 50.80. Full article

(1) One in four hospital readmissions is potentially preventable. Machine learning (ML) models have been developed to predict hospital readmissions and risk-stratify patients, but thus far they have been limited in clinical applicability, timeliness, and generalizability. (2) Methods: Using deidentified clinical data from the University of California, San Francisco (UCSF) between January 2016 and November 2021, we developed and compared four supervised ML models (logistic regression, random forest, gradient boosting, and XGBoost) to predict 30-day readmissions for adults admitted to a UCSF hospital. (3) Results: Of 147,358 inpatient encounters, 20,747 (13.9%) patients were readmitted within 30 days of discharge. The final model selected was XGBoost, which had an area under the receiver operating characteristic curve of 0.783 and an area under the precision-recall curve of 0.434. The most important features by Shapley Additive Explanations were days since last admission, discharge department, and inpatient length of stay. (4) Conclusions: We developed and internally validated a supervised ML model to predict 30-day readmissions in a US-based healthcare system. This model has several advantages including state-of-the-art performance metrics, the use of clinical data, the use of features available within 24 h of discharge, and generalizability to multiple disease states. Full article